From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: Re: [PATCH 1/5] Forking ext4 filesystem from ext3 filesystem Date: Thu, 10 Aug 2006 12:54:29 -0700 Message-ID: <20060810125429.6dded9b6.akpm@osdl.org> References: <1155172622.3161.73.camel@localhost.localdomain> <20060809233914.35ab8792.akpm@osdl.org> <44DB8036.5020706@us.ibm.com> <20060810122340.185b8d8f.akpm@osdl.org> <1155238570.12082.11.camel@kleikamp.austin.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: linux-fsdevel@vger.kernel.org, cmm@us.ibm.com, ext2-devel@lists.sourceforge.net, Badari Pulavarty , linux-kernel@vger.kernel.org Return-path: To: Dave Kleikamp In-Reply-To: <1155238570.12082.11.camel@kleikamp.austin.ibm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: ext2-devel-bounces@lists.sourceforge.net Errors-To: ext2-devel-bounces@lists.sourceforge.net List-Id: linux-fsdevel.vger.kernel.org On Thu, 10 Aug 2006 14:36:10 -0500 Dave Kleikamp wrote: > On Thu, 2006-08-10 at 12:23 -0700, Andrew Morton wrote: > > On Thu, 10 Aug 2006 11:51:34 -0700 > > Badari Pulavarty wrote: > > > > > Andrew Morton wrote: > > > > Also, JBD is presently feeding into submit_bh() buffer_heads which span two > > > > machine pages, and some device drivers spit the dummy. It'd be better to > > > > fix that once, rather than twice.. > > > > > > > Andrew, > > > > > > I looked at this few days ago. I am not sure how we end up having > > > multiple pages (especially, > > > why we end up having buffers with bh_size > pagesize) ? Do you know why ? > > > > > > > It's one or both of the jbd_kmalloc(bh->b_size) calls in > > fs/jbd/transaction.c. Here we're allocating data to attach to a bh which > > later gets fed into submit_bh(). > > > > Problem is, with CONFIG_DEBUG_SLAB=y, the data which kmalloc() returns can > > be offset by 4 bytes due to redzoning. > > > > Example: if the fs is using a 1k blocksize and we have a 4k pagesize, the > > data coming back from kmalloc may have an address of 0xnnnnxc04, so the > > data which we later feed into submit_bh() will span two pages. > > > > A simple fix would be to replace kmalloc() with a call to alloc_page(). > > We'd need to work out how much memory that will worst-case-waste. If "not > > much" then OK. > > > > If "quite a lot in the worst case" then we'd need something more elaborate. > > Would some like this be okay: > > #ifdef CONFIG_DEBUG_SLAB > return alloc_page(... > #else > return kmalloc(... > #endif > > This keeps it simple, and should be still be efficient in the > non-DEBUG_SLAB case. > I guess that would work OK. It does appear that a lot of people build and distribute CONFIG_DEBUG_SLAB kernels though. Fedora, for one. ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642